Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    specific purposes or reasons for doing something within a context

    oai_token-act-pair · gpt-3.5-turbo

    references to various purposes or uses for actions or policies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Patriarch
    -0.82
     Trees
    -0.69
    Viol
    -0.64
     Sisters
    -0.62
     Cities
    -0.62
     Oaks
    -0.62
     weeds
    -0.61
     swe
    -0.61
     Appalach
    -0.61
    ãģį
    -0.59
    POSITIVE LOGITS
    rative
    0.97
     purposes
    0.93
    farious
    0.85
    resy
    0.84
    istry
    0.80
    ptions
    0.76
    ional
    0.75
    phabet
    0.75
    oldown
    0.74
    tenance
    0.74
    Act Density 0.031%

    No Known Activations